Blog links February 15: Measuring Fraud in Russian Elections, Slum Upgrading, Causal Chains, and more…

·         Roving Bandit on the new Cochrane review on the impact of slum upgrading.

·         How do we maintain participant confidentiality in an era of open data? Discussion on the LSE Impact of Social Sciences blog.

·         Measuring voting fraud in Russia’s parliamentary elections by randomly assigning observers to monitor vote-counting and balloting – Princeton summarizes a new paper in PNAS - “According to official results, United Russia received 46.6 percent of the vote in Moscow. The results of the experiment indicate that the presence of observers led to a decrease in the share of votes for the party of 10.8 percentage points and an increase in results for other parties” (h/t @betsylevyp)

·         Chris Blattman on why you shouldn’t do a field experiment for your dissertation, and when you should ignore his advice.

·         On the 3ie blog, Howard White illustrates how tracing through the causal chain can show where your project is falling apart. He also touched on a pet peeve of mine – presenting impacts as relative impacts leading the effects to seem much larger than they are: “different ways of presenting regression models can give a misleading sense of impact. A large reduction in relative risk – a ‘good odds ratio’ – can reflect quite a small change in absolute risk. Three randomised controlled trials have found circumcision reduces the risk of transmission during unprotected sex by around 50 percent. The reduction in risk was from around 3.5 percent to 1.5 percent. Just a 2 percentage point absolute reduction, so 50 men need to be circumcised to avoid one new case of HIV/AIDS.”

·         Tom Murphy continues his annual Aid Blogger’s Best Awards (the ABBAs), and is taking nominations for favorite blogs, blog posts, etc. now – so nominate us  (or your other favorite bloggers) if you want. If you want some suggestions for favorite post of the past year, some of my favorites from DI over the past year are:

-          Markus’s post on Should we Believe the Hype About Adolescent Girls?

-          Jed’s Halloween post on the Long-run Impact of Locust Swarms

-          Berk’s post on Mind your Cowpeas and Cues: Inference and External Validity in RCTs

-          And my Father’s Day post on Dads and Development